State Estimation Filters

Precise state estimation is crucial for any autonomous system as it enables the system to provide appropriate control inputs to its actuators. However, estimation is challenging as it must be able to handle imprecise system models (often described by complex non-linear functions) as well as noisy sensor measurements in order to compute the system’s state. In this work, we explore three approaches to state estimation filtering that have different theoretical guarantees and estimation accuracies depending on the system in question (e.g., a linear system with Gaussian noise). We describe the theoretical foundations of each filter and track its evolution over time. Finally, the differences and similarities between the filters are emphasized, and some appropriate applications are discussed.